Experiment: Reactive analysis with skip-lite CMT cache

[cristianoc]

Vendor skip-lite library and integrate reactive analysis capabilities:

• Vendor skip-lite marshal_cache and reactive_file_collection modules
• Add ReactiveAnalysis module for incremental file processing
• Add CLI flags: -cmt-cache, -reactive, -runs
• Add README.md with usage and benchmark instructions

Benchmark results (~5000 files):

• Standard: CMT processing 0.78s, Total 1.01s
• Reactive (warm): CMT processing 0.01s, Total 0.20s
• Speedup: 74x for CMT processing, 5x total

The reactive mode caches processed file_data and uses read_cmt_if_changed to skip unchanged files entirely on subsequent runs.

[cristianoc]

On real project kindly provided:

=== Run 1/3 ===
Reactive: 1888 files processed, 0 from cache

=== Timing ===
  CMT processing:     2.795s (90.3%)
    - Wall clock:        2.795s
    - Result collection: 0.000ms (aggregate)
  Analysis:           0.299s (9.7%)
    - Merging:          147.725ms
    - Solving:          151.455ms
  Reporting:          0.000s (0.0%)
  Total:              3.094s

=== Run 2/3 ===
Reactive: 0 files processed, 1888 from cache

=== Timing ===
  CMT processing:     0.009s (3.7%)
    - Wall clock:        0.009s
    - Result collection: 0.000ms (aggregate)
  Analysis:           0.231s (96.3%)
    - Merging:          126.980ms
    - Solving:          103.579ms
  Reporting:          0.000s (0.0%)
  Total:              0.239s

=== Run 3/3 ===
Reactive: 0 files processed, 1888 from cache

=== Timing ===
  CMT processing:     0.008s (3.3%)
    - Wall clock:        0.008s
    - Result collection: 0.000ms (aggregate)
  Analysis:           0.248s (96.7%)
    - Merging:          139.188ms
    - Solving:          108.941ms
  Reporting:          0.000s (0.0%)
  Total:              0.257s